ABSTRACT
London forms a particularly appropriate case study on which to test the theory developed, since it has strong traditions in hospital and medical work spanning several centuries, and the historical and statistical record regarding the establishment, sizes, locations and functions of the hospital facilities are long and reasonably accurate. The problem is to provide a description of the urban population and area. If this description is too detailed, the aim of providing simple generalisations, applicable to other cities is made more difficult than if only the broad trends in population and area are evaluated. There are numerous cautionary points to note in fitting density functions to cities. The results are sensitive to the method of estimation, the areas of the population accounting units, and other factors. The problem in collecting historical data is to decide on the definition of a hospital.
